A fast lossless compression scheme for digital map images using color separation

摘要

In this paper, we present a fast lossless compression scheme for digital map images in the raster image format. This work contains two main contributions. The first is centered around the creation of a code book that is based on symbol-entropy. The second contribution is the introduction of a new row-column reduction algorithm. Our scheme proceeds as follows: we determine the number of different colors in a given map image. For each color, we create a separate bi-level data layer, one for the color and the second is for the background. Then, we compress each bi-level layer individually using the proposed method, which is based on symbol-entropy in conjunction with our row-column reduction coding algorithm. Our experimental results show that our lossless compression scheme achieved on average a compression equal to 0.035 bits per pixel which is better than most reported results in the literature or comparable to some. Moreover, our scheme is simple and fast.